[
https://issues.apache.org/jira/browse/HDFS-8005?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14503759#comment-14503759
]
Zhe Zhang commented on HDFS-8005:
---------------------------------
Good thoughts Walter. This is an important aspect we should take care of in the
decomm process. But with the current EC reconstruction logic, the selected
source replicas are not re-replicated either. They are used to reconstruct the
lost replicas and then discarded. It would be cool if we can save those
to-be-decommissioned EC replicas and push them to new destinations as well.
Let's file an optimization JIRA under HDFS-8031?
> Erasure Coding: simplify striped block recovery work computation and add tests
> ------------------------------------------------------------------------------
>
> Key: HDFS-8005
> URL: https://issues.apache.org/jira/browse/HDFS-8005
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Jing Zhao
> Assignee: Jing Zhao
> Fix For: HDFS-7285
>
> Attachments: HDFS-8005.000.patch, HDFS-8005.000.patch,
> HDFS-8005.001.patch
>
>
> HDFS-7369 adds the functionality to distribute recovery work of striped
> blocks to datanodes. There are still some pending issues:
> # In {{BlockManager#chooseSourceNode}}, a node is added into
> {{healthyIndices}} without checking if its block is live and healthy
> # The test {{TestRecoverStripedBlcoks#testMissingStripedBlock}} has not
> tested striped blocks because the file is created before setting the storage
> policy
> # In {{computeRecoveryWorkForBlocks}}, instead of using
> {{BlockCollection#isStriped}}, we'd better use {{BlockInfo#isStriped}}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
